Papers with mega-coauthorship (papers with more than 50 authors) are of specific question, IMHO. I think that 99% of authors in such papers are included “by default” since they are the directors of organizations participating in projects, they are “famous and honorary scientists” and so on, but they never wrote this publications.
Since the length of such articles is no more than 30 pages (I mean the text itself without list of coauthors and the list of references) so the contribution of each authors is maximum 25 or 30 letters in case of several hundreds (and sometimes several thousands) of coauthors. It is absurd!
Therefore the only solution here IMHO is to exclude papers with more than 15 authors (or 20 or 50, it is a question of a big discussion) from analysis of author publication activity, they should not be counted as journal publications at all, but as some “note on the experiment results” regardless of the impact factor of the journal where they are published (Reviews of Modern Physics or Physical Review Letters).
Therefore these articles should not be included in citation index, Hirsh-index and other bibliometric indicators. They may be listed on the author personal page, and in his research portfolio but no more.
In this case Author should be listed as “participant of Project A” (plus list of his 15 mega-authored articles in Reviews of Modern Physics), “project B research manager” (plus list of his 10 mega-authored articles in Physical Review Letters) and so on.
I do not say here that mega-authored papers are of bad quality or something wrong. Absolutely not!
My idea the following. Publication is (to some extent in case of multi-author publications) the form of INDIVIDUAL work. Therefore, there is obviously some limit on the number of coauthors for clear detecting the INDIVIDUAL contribution of EACH coauthor. If the number of authors exceeds 20 (or even less) can we CLEARLY DETECT THE INDIVIDUAL CONTRIBUTION at all regardless of the length of paper regardless its scientific level and the level of impact factor of journal where it is published?
When the number of authors exceeds the reasonable level in my opinion this paper cannot be treated as scientific article.
Let me take the real example.
Article title “Search for quark contact interactions in dijet angular distributions in pp collisions at root s=7 TeV measured with the ATLAS detector”
Journal: NEW JOURNAL OF PHYSICS
Number of authors – 3178 (!!!)
Number of pages with text only: 23 (A4 format of the sheet; Times New Roman, 12; line spacing 1)
Number of cites (citations) in Web of Science: 43 (by 23 March 2013)
Number of letters – 48000 approx. Each authors contribution 16 (!!!) letters. Not pages, not paragraphs, not even words, only 16 LETTERS.
INDIVIDUAL contribution for EACH AUTHOR will look something like this “As a cross-check, a B” or “Finally, an analys”, “distributions, pla” and so on.
Can we treat this item as AN INDIVIDUAL ARTICLE???? Absolutely no! Never!
The other question: where was the editorial board when peer-reviewing and publishing this “publication”????
What is your opinions on my ideas?
Citation analysis and bibliometrics has extended to the web to include webometrics. This increases the reach of academic publishing beyond the traditional world of hard-copy journals, peer-reviewed articles, books, book chapters, etc. as the sources of scholarly evaluation. Other scholarly products including, conference presentations, proceedings, research reports, etc. are being accessed by not only academics but also the general public.